According to new research from the National Institute of Standards and Technology (NIST), even the best facial recognition algorithms are tricked by the use of face masks. Face masks have become increasingly popular, and mandatory in various stores and businesses as the US works to combat the spread of COVID-19, however, the masks have also proven successful in deterring facial recognition technology from identifying people.
Even sophisticated facial recognition systems clocked error rates between 5 and 50 percent when matching people in digitally applied face masks with photos of the same person without a mask. The study shows the limitations of facial recognition technology, particularly in a post-pandemic world. The research aims to develop and test new technology that assesses how to identify masked individuals.